Computational Cognitive Model Research Articles

Learning to drive: A reconceptualization

Drivers’ population-level crash rates incrementally decrease following licensure, which has led to the implicit assumption that an individual driver’s crash risk also decreases incrementally after licensure as they accrue experience. However, in the aggregate data an incremental decrease in crash rate can reflect both incremental reductions in crash risk within individuals and an incremental increase in the proportion of drivers who have experienced an abrupt decrease in crash risk. Therefore, while it is true to say that the population of drivers’ crash risk reduces in the months following licensure, it is not necessarily true to say that a driver’s crash risk reduces in the months following licensure; that is, it cannot be assumed that individual-level changes in crash risk mirror the population-level changes in crash rates. In statistics, this is known as an ecological fallacy and in formal logic it is known as the fallacy of division, a type of category error. Using computational cognitive modeling methods we demonstrate that aggregating individual-level abrupt decreases in crash risk (i.e., non-incremental change trajectories) accurately fits population-level crash rate data from over 1 million novice drivers and uniquely accounts for effects of two interventions found to reduce police-reported MVCs. Thus, we demonstrate that: (1) a power-law artifact is readily observable in newly licensed drivers’ aggregate crash data, which is not necessarily indicative of individual-level change processes, (2) interventions can alter crash risk trajectories by inducing immediate phase changes in crash risk into a lower risk stratum, or increasing the probability of such a change, and (3) a phase transition model provides a stronger and more parsimonious account of the existing data than an incremental-accrual model.

Suicidal thoughts and behaviors are associated with an increased decision-making bias for active responses to escape aversive states.

Suicide is a leading cause of death worldwide. Despite decades of clinical and theoretical accounts that suggest that suicidal thoughts and behaviors are efforts to escape painful emotions, little prior research has examined decision making involved in escaping aversive states. We compared the performance of 85 suicidal participants to 44 nonsuicidal psychiatric patients on a novel reinforcement learning task with choices to make either active (i.e., "go") or passive responses (i.e., "no-go") to either escape or avoid an aversive stimulus. We used a computational cognitive model to isolate decision-making biases. We hypothesized that suicidal participants would exhibit a relatively elevated bias for making active responses to escape an aversive state and would show worse performance when escape required a passive response (i.e., "doing nothing" to escape). Our hypotheses were supported: The computational model revealed that suicidal participants exhibited a higher bias for an active response to escape compared with nonsuicidal psychiatric controls, suggesting that this finding was not just the result of the presence of psychopathology. The bias parameter also accounted for unique variance in predicting group status among several constructs previously related to suicidal thoughts and behaviors. This study provides a new method for testing escape decision making and does so using a computational cognitive model, allowing us to precisely index processes underlying suicidal and related behaviors. Future research examining escape decision making from a computational perspective could help link neural processes or environmental stressors to suicidal thoughts or behaviors. (PsycINFO Database Record (c) 2019 APA, all rights reserved).

Does direction matter? Linguistic asymmetries reflected in visual attention

Language and vision interact in non-trivial ways. Linguistically, spatial utterances are often asymmetrical as they relate more stable objects (reference objects) to less stable objects (located objects). Researchers have claimed that such linguistic asymmetry should also be reflected in the allocation of visual attention when people process a depicted spatial relation described by spatial language. More specifically, it was assumed that people move their attention from the reference object to the located object. However, recent theoretical and empirical findings challenge the directionality of this attentional shift. In this article, we present the results of an empirical study based on predictions generated by computational cognitive models implementing different directionalities of attention. Moreover, we thoroughly analyze the computational models. While our results do not favor any of the implemented directionalities of attention, we found that two unknown sources of geometric information affect spatial language understanding. We provide modifications to the computational models that substantially improve their performance on empirical data.

Book review: Computational modeling of cognition and behavior

Book review: Computational modeling of cognition and behavior

How] Can Pluralist Approaches to Computational Cognitive Modeling of Human Needs and Values Save our Democracies?

[Comment] Est-ce que les approches pluralistes à la modélisation cognitive computationnelle des besoins et valeurs humains peuvent sauver nos démocraties ? Dans nos sociétés de plus en plus digitales, un grand nombre des entreprises implémentent un modèle commercial qui perçoive les données personnelles des usagers (ou des consommateurs) en tant que ressource cloisonnée, détenue et contrôlée par le contrôleur des données plutôt que les personnes concernées. La collecte et le traitement d'une telle quantité de données personnelles comporte de nombreuses conséquences techniques, sociales et économiques négatives, y compris l'atteinte à la vie privée des personnes et à l'autonomie des individus. En conséquence, de nouvelles réglementations telles que le règlement. général européen sur la protection des données (RGPD) ont tenté de prendre des mesures importantes pour une meilleure mise en oeuvre du droit à la vie privée digitale. Le présent document propose que ces actes juridiques s'accompagnent du développement de solutions techniques complémentaires telles que les systèmes d''assistants personnels cognitifs qui pourraient aider les personnes à gérer le traitement de leurs données personnelles sur Internet. Considérant l'importance et la sensibilité du traitement des données à caractère personnel, ces systèmes assistants devraient non seulement tenir compte des besoins et des valeurs de leurs propriétaires, mais aussi être transparents, responsables et contrôlables. Les approches pluralistes dans la modélisation cognitive computationnelle des besoins et des valeurs humaines qui ne sont pas limitées par des frontières paradigmatiques traditionnelles comme le cognitivisme, le connectionnisme ou l'énactivisme, pourraient créer un équilibre entre l'aspect pratique et l'utilité, d'une part, et la transparence, la responsabilité et le contrôle, d'autre part, tout en appuyant et en donnant du pouvoir aux humains dans le monde digital. En considérant la menace pour la vie privée digitale comme une menace significative pour les démocraties contemporaines, la mise en oeuvre future de tels modèles pluralistes contribuerait à l'équilibre des pouvoirs, l'équité et l'inclusion dans nos sociétés.

Emergentist View on Generative Narrative Cognition: Considering Principles of the Self-Organization of Mental Stories

We consider the essence of human intelligence to be the ability to mentally (internally) construct a world in the form of stories through interactions with external environments. Understanding the principles of this mechanism is vital for realizing a human-like and autonomous artificial intelligence, but there are extremely complex problems involved. From this perspective, we propose a conceptual-level theory for the computational modeling of generative narrative cognition. Our basic idea can be described as follows: stories are representational elements forming an agent’s mental world and are also living objects that have the power of self-organization. In this study, we develop this idea by discussing the complexities of the internal structure of a story and the organizational structure of a mental world. In particular, we classify the principles of the self-organization of a mental world into five types of generative actions, i.e., connective, hierarchical, contextual, gathering, and adaptive. An integrative cognition is explained with these generative actions in the form of a distributed multiagent system of stories.

Una aproximación conexionista a los procesos mentales. Entrevista con James L. McClelland

In this interview, James L. McClelland responds to questions regarding connectionist models of cognition, a theory inspired by information processing in the brain. McClelland explains the distinction between symbolic and non-symbolic processing for a better understanding of mental processes. He argues that connectionist models can perform the computations which we know the brain can perform. In addition, he responds to several general questions on the perspectives of computational models of cognition.

Explaining Explanation For “Explainable Ai”

What makes for an explanation of “black box” AI systems such as Deep Nets? We reviewed the pertinent literatures on explanation and derived key ideas. This set the stage for our empirical inquiries, which include conceptual cognitive modeling, the analysis of a corpus of cases of "naturalistic explanation" of computational systems, computational cognitive modeling, and the development of measures for performance evaluation. The purpose of our work is to contribute to the program of research on “Explainable AI.” In this report we focus on our initial synthetic modeling activities and the development of measures for the evaluation of explainability in human-machine work systems.

Developing and Evaluating an Augmented Reality Interface to Assist the Joint Tactical Air Controller by Applying Human Performance Models

We developed a 3D augmented reality head mounted display (DARSADS-SVS HMD) interface to support the Joint Tactical Air Controller (JTAC). The JTAC’s job is to integrate information about enemy attack units and nearby friendly forces and direct aircraft equipped with weapons to neutralize the enemy via close air support (CAS), while also safely routing air traffic. The JTAC’s numerous and often overlapping tasks involve maintaining detailed situational awareness (SA) of a large quantity of information, and making rapid decisions that carry life-or-death consequences. Thus, the JTAC role requires many different cognitive operations across different mission phases. Designing an effective human-factored system that supports maximum SA while minimizing cognitive load required us to harness computational cognitive models of SA-supporting visual scanning, display layout, 3D frame-of-reference transformations, clutter, legibility and working memory. We applied such models to different phases of the JTAC mission (e.g., airspace management, call-for-fire), establishing a Figure of Merit (FOM) for each given design by summing FOMs across models, thus creating a mechanism to evaluate designs based upon their balanced impact on competing cognitive drivers. Models were differentially weighted for each phase, according to the relative importance of the relevant cognitive process to the phase in question. In this research paper, we illustrate two such design comparisons.

Interfacing the Modifiable Multitasking Environment with ACT-R for Computational Cognitive Modeling of Complex Tasks

We demonstrate a set of software tools designed to facilitate computational cognitive modeling of multitasking performance. The Modifiable Multitasking Environment (ModME) offers a flexible, browser-based platform for creating multitasking experiments. Simplified Interfacing for Modeling Cognition–JavaScript (SIMCog-JS) provides communication between the browser-based experiments in ModME and the Java implementation of the ACT-R cognitive architecture. The baseline configuration of these software packages enables an ACT-R model to perform pilot-like multitasking in the modified Multi-Attribute Task Battery, which is implemented as the baseline task available in ModME. We show how this combination facilitates the development of models for assessing multitasking workload. In this demonstration, we will explain the software packages and allow attendees to interact with system elements, particularly the ModME graphical user interfaces. All software is available open source for attendees to try themselves.

Learning features while learning to classify: a cognitive model for autonomous systems

We describe a computational cognitive model intended to be a generalizable classifier that can provide context-based feedback to semantic perception in robotic applications. Many classifiers (including cognitive models of categorization) perform well at the task of associating features with objects. Underlying their performance is an effective selection of the features used during classification. This feature selection (FS) process is usually performed outside the boundaries of the models that learn and perform classification tasks, often by human experts. In contrast, the cognitive model we describe simultaneously learns which features to use, as it learns the associations between features and classes. This integration of FS and class learning in one model makes it complementary to other machine-learning techniques that generate feature-based representations (e.g., deep learning methods). But their integration in a cognitive architecture also provides a means for creating a dynamic context that includes disparate sources of information (e.g., environmental observations, task knowledge, commands from humans). This richer context, in turn, provides a means for making semantic perception goal-directed. We demonstrate automated FS, integrated with an instance-based learning approach to classification, in an ACT-R model of categorization by labeling facial expressions of emotion (e.g., happy, sad), and then generalizing the model to the classification of indoor public spaces (e.g., cafes, classrooms).

An Automated Method for Building Cognitive Models for Turn-Based Games from a Strategy Logic

Whereas game theorists and logicians use formal methods to investigate ideal strategic behavior, many cognitive scientists use computational cognitive models of the human mind to predict and simulate human behavior. In this paper, we aim to bring these fields closer together by creating a generic translation system which, starting from a strategy for a turn-based game represented in formal logic, automatically generates a computational model in the Primitive Information Processing Elements (PRIMs) cognitive architecture, which has been validated on various experiments in cognitive psychology. The PRIMs models can be run and fitted to participants’ data in terms of decisions, response times, and answers to questions. As a proof of concept, we run computational modeling experiments on the basis of a game-theoretic experiment about the turn-based game “Marble Drop with Surprising Opponent”, in which the opponent often starts with a seemingly irrational move. We run such models starting from logical representations of several strategies, such as backward induction and extensive-form rationalizability, as well as different player types according to stance towards risk and level of theory of mind. Hereby, response times and decisions for such centipede-like games are generated, which in turn leads to concrete predictions for future experiments with human participants. Such precise predictions about different aspects, including reaction times, eye movements and active brain areas, cannot be derived on the basis of a strategy logic by itself: the computational cognitive models play a vital role and our generic translation system makes their construction more efficient and systematic than before.

A computational model of cognitive development for the motor skill learning from curiosity

Abstract Curiosity is a key enabler of autonomous learning for cognitive development of humans. It allows them to learn motor skills through autonomous exploration of the environment. This paper proposes a new neurobiologically-inspired cognitive computational model: C-DCCM (Curiosity-Driven Cognitive Computing Model). The objective of this model is to implement the effect of curiosity that can be applied to the balance learning problem of the two-wheeled self-balancing robot. As basis of the modeling we use and extend the Actor-Critic model. Verification is conducted via simulation on a computing system of sustained dopamine modulation mechanism and its effects on the cerebral cortex. In order to illustrate the effect of the curiosity on the autonomous exploration, some comparison experiments about the balance learning problem have been done.

A Generative Model of Cognitive State from Task and Eye Movements.

The early eye tracking studies of Yarbus provided descriptive evidence that an observer's task influences patterns of eye movements, leading to the tantalizing prospect that an observer's intentions could be inferred from their saccade behavior. We investigate the predictive value of task and eye movement properties by creating a computational cognitive model of saccade selection based on instructed task and internal cognitive state using a Dynamic Bayesian Network (DBN). Understanding how humans generate saccades under different conditions and cognitive sets links recent work on salience models of low-level vision with higher level cognitive goals. This model provides a Bayesian, cognitive approach to top-down transitions in attentional set in pre-frontal areas along with vector-based saccade generation from the superior colliculus. Our approach is to begin with eye movement data that has previously been shown to differ across task. We first present an analysis of the extent to which individual saccadic features are diagnostic of an observer's task. Second, we use those features to infer an underlying cognitive state that potentially differs from the instructed task. Finally, we demonstrate how changes of cognitive state over time can be incorporated into a generative model of eye movement vectors without resorting to an external decision homunculus. Internal cognitive state frees the model from the assumption that instructed task is the only factor influencing observers' saccadic behavior. While the inclusion of hidden temporal state does not improve the classification accuracy of the model, it does allow accurate prediction of saccadic sequence results observed in search paradigms. Given the generative nature of this model, it is capable of saccadic simulation in real time. We demonstrated that the properties from its generated saccadic vectors closely match those of human observers given a particular task and cognitive state. Many current models of vision focus entirely on bottom-up salience to produce estimates of spatial "areas of interest" within a visual scene. While a few recent models do add top-down knowledge and task information, we believe our contribution is important in three key ways. First, we incorporate task as learned attentional sets that are capable of self-transition given only information available to the visual system. This matches influential theories of bias signals by (Miller and Cohen Annu Rev Neurosci 24:167-202, 2001) and implements selection of state without simply shifting the decision to an external homunculus. Second, our model is generative and capable of predicting sequence artifacts in saccade generation like those found in visual search. Third, our model generates relative saccadic vector information as opposed to absolute spatial coordinates. This matches more closely the internal saccadic representations as they are generated in the superior colliculus.

Thinking in PolAR Pictures: Using Rotation-Friendly Mental Images to Solve Leiter-R Form Completion

The Leiter International Performance Scale-Revised (Leiter-R) is a standardized cognitive test that seeks to "provide a nonverbal measure of general intelligence by sampling a wide variety of functions from memory to nonverbal reasoning." Understanding the computational building blocks of nonverbal cognition, as measured by the Leiter-R, is an important step towards understanding human nonverbal cognition, especially with respect to typical and atypical trajectories of child development. One subtest of the Leiter-R, Form Completion, involves synthesizing and localizing a visual figure from its constituent slices. Form Completion poses an interesting nonverbal problem that seems to combine several aspects of visual memory, mental rotation, and visual search. We describe a new computational cognitive model that addresses Form Completion using a novel, mental-rotation-friendly image representation that we call the Polar Augmented Resolution (PolAR) Picture, which enables high-fidelity mental rotation operations. We present preliminary results using actual Leiter-R test items and discuss directions for future work.

Bridging Behavioral and Naturalistic Decision-Making Research by Computational Cognitive Models

Bridging Behavioral and Naturalistic Decision-Making Research by Computational Cognitive Models

Applying Cognitive Learner Models for Recommender Systems in Sparse Data Learning Environments

In recent years, various recommendation algorithms have been proposed to support learners in technology-enhanced learning environments. Such algorithms have proven to be quite effective in big-data learning settings (massive open online courses), yet successful applications in other informal and formal learning settings are rare. Common challenges include data sparsity, the lack of sufficiently flexible learner and domain models, and the difficulty of including pedagogical goals into recommendation strategies. Computational models of human cognition and learning are, in principle, well positioned to help meet these challenges, yet the effectiveness of cognitive models in educational recommender systems remains poorly understood to this date. This thesis contributes to this strand of research by investigating i) two cognitive learner models (CbKST and SUSTAIN) for resource recommendations that qualify for sparse user data by following theory-driven top down approaches, and ii) two tag recommendation strategies based on models of human cognition (BLL and MINERVA2) that support the creation of learning content meta-data. The results of four online and offline experiments in different learning contexts indicate that a recommendation approach based on the CbKST, a well-founded structural model of knowledge representation, can improve the users? perceived learning experience in formal learning settings. In informal settings, SUSTAIN, a human category learning model, is shown to succeed in representing dynamic, interest based learning interactions and to improve Collaborative Filtering for resource recommendations. The investigation of the two proposed tag recommender strategies underlined their ability to generate accurate suggestions (BLL) and in collaborative settings, their potential to promote the development of shared vocabulary (MINERVA2). This thesis shows that the application of computational models of human cognition holds promise for the design of recommender mechanisms and, at the same time, for gaining a deeper understanding of interaction dynamics in virtual learning systems.

How Language Processing can Shape a Common Model of Cognition

What role does the study of natural language play in the task of developing a unified theory and common model of cognition? Language is perhaps the most complex behaviour that humans exhibit, and, as such, is one of the most difficult problems for understanding human cognition. Linguistic theory can both inform and be informed by unified models of cognition. We discuss (1) how computational models of human cognition can provide insight into how humans produce and comprehend language and (2) how the problem of modelling language processing raises questions and creates challenges for widely used computational models of cognition. Evidence from the literature suggests that behavioural phenomena, such as recency and priming effects, and cognitive constraints, such as working memory limits, affect how language is produced by humans in ways that can be predicted by computational cognitive models. But just as computational models can provide new insights into language, language can serve as a test for these models. For example, simulating language learning requires the use of more powerful machine learning techniques, such as deep learning and vector symbolic architectures, and language comprehension requires a capacity for on-the-fly situational model construction. In sum, language plays an important role in both shaping the development of a common model of the mind, and, in turn, the theoretical understanding of language stands to benefit greatly from the development of a common model.

The wandering self: Tracking distracting self-generated thought in a cognitively demanding context.

We investigated how self-referential processing (SRP) affected self-generated thought in a complex working memory task (CWM) to test the predictions of a computational cognitive model. This model described self-generated thought as resulting from competition between task- and distracting processes, and predicted that self-generated thought interferes with rehearsal, reducing memory performance. SRP was hypothesized to influence this goal competition process by encouraging distracting self-generated thinking. We used a spatial CWM task to examine if SRP instigated such thoughts, and employed eye-tracking to examine rehearsal interference in eye-movement and self-generated thinking in pupil size. The results showed that SRP was associated with lower performance and higher rates of self-generated thought. Self-generated thought was associated with less rehearsal and we observed a smaller pupil size for mind wandering. We conclude that SRP can instigate self-generated thought and that goal competition provides a likely explanation for how self-generated thoughts arises in a demanding task.

Cognitive computing and eScience in health and life science research: artificial intelligence and obesity intervention programs.

To present research models based on artificial intelligence and discuss the concept of cognitive computing and eScience as disruptive factors in health and life science research methodologies. The paper identifies big data as a catalyst to innovation and the development of artificial intelligence, presents a framework for computer-supported human problem solving and describes a transformation of research support models. This framework includes traditional computer support; federated cognition using machine learning and cognitive agents to augment human intelligence; and a semi-autonomous/autonomous cognitive model, based on deep machine learning, which supports eScience. The paper provides a forward view of the impact of artificial intelligence on our human-computer support and research methods in health and life science research. By augmenting or amplifying human task performance with artificial intelligence, cognitive computing and eScience research models are discussed as novel and innovative systems for developing more effective adaptive obesity intervention programs.